From: Richard Henderson <richard.henderson@linaro.org>
To: Evan Green <evan@rivosinc.com>
Cc: Florian Weimer <fweimer@redhat.com>,
libc-alpha@sourceware.org, slewis@rivosinc.com,
palmer@rivosinc.com, vineetg@rivosinc.com
Subject: Re: [PATCH v6 5/5] riscv: Add and use alignment-ignorant memcpy
Date: Thu, 3 Aug 2023 15:30:27 -0700 [thread overview]
Message-ID: <548fc7d5-6225-69e7-f4a7-47669d2fdbd5@linaro.org> (raw)
In-Reply-To: <CALs-HssWPZoMpuqxfEg3+bVdRgvm03ZYBPrAAjKXx=ojsoemYw@mail.gmail.com>
On 8/3/23 11:42, Evan Green wrote:
> On Thu, Aug 3, 2023 at 10:50 AM Richard Henderson
> <richard.henderson@linaro.org> wrote:
>> Outside libc something is required.
>>
>> An extra parameter to ifunc is surprising though, and clearly not ideal per the extra
>> hoops above. I would hope for something with hidden visibility in libc_nonshared.a that
>> could always be called directly.
>
> My previous spin took that approach, defining a
> __riscv_hwprobe_early() in libc_nonshared that could route to the real
> function if available, or make the syscall directly if not. But that
> approach had the drawback that ifunc users couldn't take advantage of
> the vDSO, and then all users had to comprehend the difference between
> __riscv_hwprobe() and __riscv_hwprobe_early().
I would define __riscv_hwprobe such that it could take advantage of the vDSO once
initialization reaches a certain point, but cope with being run earlier than that point by
falling back to the syscall.
That constrains the implementation, I guess, in that it can't set errno, but just
returning the negative errno from the syscall seems fine.
It might be tricky to get a reference to GLRO(dl_vdso_riscv_hwprobe) very early, but I
would hope that some application of __attribute__((weak)) might correctly get you a NULL
prior to full relocations being complete.
> In contrast, IMO this approach is much nicer. Ifunc writers are
> already used to getting hwcap info via a parameter. Adding this second
> parameter, which also provides hwcap-like things, seems like a natural
> extension. I didn't quite follow what you meant by the "extra hoops
> above".
The check for null function pointer, for sure. But also consider how __riscv_hwprobe is
going to be used.
It might be worth defining some helper functions for probing a single key or a single
field. E.g.
uint64_t __riscv_hwprobe_one_key(int64_t key, unsigned int flags)
{
struct riscv_hwprobe pair = { .key = key };
int err = __riscv_hwprobe(&pair, 1, 0, NULL, flags);
if (err)
return err;
if (pair.key == -1)
return -ENOENT;
return pair.value;
}
This implementation requires that no future hwprobe key define a value which as a valid
value in the errno range (or better, bit 63 unused). Alternately, or additionally:
bool __riscv_hwprobe_one_mask(int64_t key, uint64_t mask, uint64_t val, int flags)
{
struct riscv_hwprobe pair = { .key = key };
return (__riscv_hwprobe(&pair, 1, 0, NULL, flags) == 0
&& pair.key != -1
&& (pair.value & mask) == val);
}
These yield either
int64_t v = __riscv_hwprobe_one_key(CPUPERF_0, 0);
if (v >= 0 && (v & MISALIGNED_MASK) == MISALIGNED_FAST)
return __memcpy_noalignment;
return __memcpy_generic;
or
if (__riscv_hwprobe_one_mask(CPUPERF_0, MISALIGNED_MASK, MISALIGNED_FAST, 0))
return __memcpy_noalignment;
return __memcpy_generic;
which to my mind looks much better for a pattern you'll be replicating so very many times
across all of the ifunc implementations in the system.
r~
next prev parent reply other threads:[~2023-08-03 22:30 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-02 15:58 [PATCH v6 0/5] RISC-V: ifunced memcpy using new kernel hwprobe interface Evan Green
2023-08-02 15:58 ` [PATCH v6 1/5] riscv: Add Linux hwprobe syscall support Evan Green
2023-08-02 16:52 ` Joseph Myers
2023-08-03 7:24 ` Florian Weimer
2023-08-02 15:59 ` [PATCH v6 2/5] riscv: Add hwprobe vdso call support Evan Green
2023-08-02 15:59 ` [PATCH v6 3/5] riscv: Add __riscv_hwprobe pointer to ifunc calls Evan Green
2023-08-02 15:59 ` [PATCH v6 4/5] riscv: Enable multi-arg ifunc resolvers Evan Green
2023-08-02 15:59 ` [PATCH v6 5/5] riscv: Add and use alignment-ignorant memcpy Evan Green
2023-08-03 7:25 ` Florian Weimer
2023-08-03 17:50 ` Richard Henderson
2023-08-03 18:42 ` Evan Green
2023-08-03 22:30 ` Richard Henderson [this message]
2023-08-07 22:10 ` Evan Green
2023-08-07 22:21 ` Florian Weimer
2023-08-07 22:30 ` Evan Green
2023-08-07 22:48 ` enh
2023-08-08 0:01 ` Evan Green
2023-08-12 0:01 ` enh
2023-08-15 16:40 ` Evan Green
2023-08-15 21:53 ` enh
2023-08-15 23:01 ` Evan Green
2023-08-16 23:18 ` enh
2023-08-17 16:27 ` Evan Green
2023-08-17 16:37 ` enh
2023-08-17 17:40 ` Evan Green
2023-08-22 15:06 ` enh
2023-08-02 16:03 ` [PATCH v6 0/5] RISC-V: ifunced memcpy using new kernel hwprobe interface Evan Green
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=548fc7d5-6225-69e7-f4a7-47669d2fdbd5@linaro.org \
--to=richard.henderson@linaro.org \
--cc=evan@rivosinc.com \
--cc=fweimer@redhat.com \
--cc=libc-alpha@sourceware.org \
--cc=palmer@rivosinc.com \
--cc=slewis@rivosinc.com \
--cc=vineetg@rivosinc.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).